paper_random_eval.jsonl

{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:42.620987+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.200304985046387, "chaser_new_fields": 80.66999816894531, "chaser_partner_vision": 0.0, "chaser_return": 5.999996185302734, "chaser_stuck_run_steps": 0.019999999552965164, "collisions_per_episode": 2.3999998569488525, "degenerate_episodes": 84, "degenerate_fraction": 0.84, "episodes": 100, "explorer_new_fields": 32.79999923706055, "explorer_partner_vision": 0.0, "explorer_return": -1.3700000047683716, "explorer_stuck_run_steps": 2.669999837875366, "final_distance": 5.008528709411621, "same_state_run_steps": 0.0}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0000"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:42.718185+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.077332973480225, "chaser_new_fields": 33.65999984741211, "chaser_partner_vision": 0.0, "chaser_return": -3.331998825073242, "chaser_stuck_run_steps": 2.700000047683716, "collisions_per_episode": 2.2699999809265137, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 74.5999984741211, "explorer_partner_vision": 0.0, "explorer_return": 61.10499954223633, "explorer_stuck_run_steps": 0.07999999821186066, "final_distance": 5.209523677825928, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0000"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:42.818944+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.202634334564209, "chaser_new_fields": 78.47000122070312, "chaser_partner_vision": 0.0, "chaser_return": 5.575997352600098, "chaser_stuck_run_steps": 0.019999999552965164, "collisions_per_episode": 2.309999942779541, "degenerate_episodes": 83, "degenerate_fraction": 0.83, "episodes": 100, "explorer_new_fields": 32.97999954223633, "explorer_partner_vision": 0.0, "explorer_return": -1.0699999332427979, "explorer_stuck_run_steps": 2.7100000381469727, "final_distance": 5.011303424835205, "same_state_run_steps": 0.0}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0001"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:42.915491+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.171538352966309, "chaser_new_fields": 33.95000076293945, "chaser_partner_vision": 0.0, "chaser_return": -3.2539987564086914, "chaser_stuck_run_steps": 2.6599998474121094, "collisions_per_episode": 2.0899999141693115, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 71.5, "explorer_partner_vision": 0.0, "explorer_return": 56.59000015258789, "explorer_stuck_run_steps": 0.3100000023841858, "final_distance": 4.9595184326171875, "same_state_run_steps": 0.05999999865889549}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0001"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00015983218477721994, "chaser_grad_norm": 0.06889200676232576, "chaser_new_fields": 84.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.7600016593933105, "chaser_subspace_norm": 98.9588623046875, "collisions_per_episode": 2.075000047683716, "entropy": 0.02094564540311694, "explorer_grad_norm": 1.1887003779411316, "explorer_new_fields": 82.75, "explorer_partner_vision": 0.0, "explorer_return": 73.30000305175781, "explorer_subspace_norm": 38.70703887939453, "final_distance": 4.667576789855957, "max_param_abs": 1.5199339389801025, "policy_loss": -0.0006238509232616707, "value_loss": 3.509905695915222}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.015998+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.099634170532227, "chaser_new_fields": 81.38999938964844, "chaser_partner_vision": 0.0, "chaser_return": 6.113996505737305, "chaser_stuck_run_steps": 0.009999999776482582, "collisions_per_episode": 3.049999952316284, "degenerate_episodes": 84, "degenerate_fraction": 0.84, "episodes": 100, "explorer_new_fields": 32.849998474121094, "explorer_partner_vision": 0.0, "explorer_return": -1.3849999904632568, "explorer_stuck_run_steps": 2.700000047683716, "final_distance": 4.918816566467285, "same_state_run_steps": 0.0}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0002"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00015983218477721994, "chaser_grad_norm": 0.06889200676232576, "chaser_new_fields": 84.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.7600016593933105, "chaser_subspace_norm": 98.9588623046875, "collisions_per_episode": 2.075000047683716, "entropy": 0.02094564540311694, "explorer_grad_norm": 1.1887003779411316, "explorer_new_fields": 82.75, "explorer_partner_vision": 0.0, "explorer_return": 73.30000305175781, "explorer_subspace_norm": 38.70703887939453, "final_distance": 4.667576789855957, "max_param_abs": 1.5199339389801025, "policy_loss": -0.0006238509232616707, "value_loss": 3.509905695915222}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.112222+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.199133396148682, "chaser_new_fields": 33.43000030517578, "chaser_partner_vision": 0.0, "chaser_return": -3.3739986419677734, "chaser_stuck_run_steps": 2.609999895095825, "collisions_per_episode": 2.049999952316284, "degenerate_episodes": 85, "degenerate_fraction": 0.85, "episodes": 100, "explorer_new_fields": 80.68999481201172, "explorer_partner_vision": 0.0, "explorer_return": 70.2699966430664, "explorer_stuck_run_steps": 0.7799999713897705, "final_distance": 4.922343730926514, "same_state_run_steps": 0.08999999612569809}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0002"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0002841827333717878, "chaser_grad_norm": 0.04017048981040716, "chaser_new_fields": 87.07500457763672, "chaser_partner_vision": 0.0, "chaser_return": 7.300001621246338, "chaser_subspace_norm": 112.96394348144531, "collisions_per_episode": 1.725000023841858, "entropy": 0.024153232108801603, "explorer_grad_norm": 0.8466866165399551, "explorer_new_fields": 75.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 62.125, "explorer_subspace_norm": 36.111541748046875, "final_distance": 4.144174098968506, "max_param_abs": 1.6294050216674805, "policy_loss": -0.0005663236785924397, "value_loss": 3.4905534386634827}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.212404+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.134275913238525, "chaser_new_fields": 86.43999481201172, "chaser_partner_vision": 0.0, "chaser_return": 7.12199592590332, "chaser_stuck_run_steps": 0.019999999552965164, "collisions_per_episode": 2.6399998664855957, "degenerate_episodes": 84, "degenerate_fraction": 0.84, "episodes": 100, "explorer_new_fields": 32.790000915527344, "explorer_partner_vision": 0.0, "explorer_return": -1.2799999713897705, "explorer_stuck_run_steps": 2.6499998569488525, "final_distance": 4.301638603210449, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0003"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0002841827333717878, "chaser_grad_norm": 0.04017048981040716, "chaser_new_fields": 87.07500457763672, "chaser_partner_vision": 0.0, "chaser_return": 7.300001621246338, "chaser_subspace_norm": 112.96394348144531, "collisions_per_episode": 1.725000023841858, "entropy": 0.024153232108801603, "explorer_grad_norm": 0.8466866165399551, "explorer_new_fields": 75.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 62.125, "explorer_subspace_norm": 36.111541748046875, "final_distance": 4.144174098968506, "max_param_abs": 1.6294050216674805, "policy_loss": -0.0005663236785924397, "value_loss": 3.4905534386634827}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.308901+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.156155109405518, "chaser_new_fields": 33.97999954223633, "chaser_partner_vision": 0.0, "chaser_return": -3.2619988918304443, "chaser_stuck_run_steps": 2.619999885559082, "collisions_per_episode": 1.8600000143051147, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 74.72000122070312, "explorer_partner_vision": 0.0, "explorer_return": 61.44999694824219, "explorer_stuck_run_steps": 0.05999999865889549, "final_distance": 5.338278293609619, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0003"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00037036755928841064, "chaser_grad_norm": 0.08058446925133467, "chaser_new_fields": 82.17500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.330001354217529, "chaser_subspace_norm": 92.44800567626953, "collisions_per_episode": 1.8250000476837158, "entropy": 0.04575740918517113, "explorer_grad_norm": 0.3279200941324234, "explorer_new_fields": 77.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.0999984741211, "explorer_subspace_norm": 29.62271499633789, "final_distance": 5.308696269989014, "max_param_abs": 1.3726928234100342, "policy_loss": -0.0006299224167634065, "value_loss": 3.322889506816864}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.409994+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.038296699523926, "chaser_new_fields": 78.72000122070312, "chaser_partner_vision": 0.0, "chaser_return": 5.573996543884277, "chaser_stuck_run_steps": 0.04999999701976776, "collisions_per_episode": 2.7899999618530273, "degenerate_episodes": 85, "degenerate_fraction": 0.85, "episodes": 100, "explorer_new_fields": 32.939998626708984, "explorer_partner_vision": 0.0, "explorer_return": -1.25, "explorer_stuck_run_steps": 2.700000047683716, "final_distance": 4.791200637817383, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0004"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00037036755928841064, "chaser_grad_norm": 0.08058446925133467, "chaser_new_fields": 82.17500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.330001354217529, "chaser_subspace_norm": 92.44800567626953, "collisions_per_episode": 1.8250000476837158, "entropy": 0.04575740918517113, "explorer_grad_norm": 0.3279200941324234, "explorer_new_fields": 77.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.0999984741211, "explorer_subspace_norm": 29.62271499633789, "final_distance": 5.308696269989014, "max_param_abs": 1.3726928234100342, "policy_loss": -0.0006299224167634065, "value_loss": 3.322889506816864}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.506454+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.115623474121094, "chaser_new_fields": 33.89999771118164, "chaser_partner_vision": 0.0, "chaser_return": -3.289998769760132, "chaser_stuck_run_steps": 2.6499998569488525, "collisions_per_episode": 2.919999837875366, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 76.08000183105469, "explorer_partner_vision": 0.0, "explorer_return": 63.32499694824219, "explorer_stuck_run_steps": 0.07000000029802322, "final_distance": 5.588272571563721, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0004"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00022429173202174373, "chaser_grad_norm": 0.04249562695622444, "chaser_new_fields": 80.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.015000820159912, "chaser_subspace_norm": 93.73216247558594, "collisions_per_episode": 1.024999976158142, "entropy": 0.05529557354748249, "explorer_grad_norm": 0.39152009785175323, "explorer_new_fields": 73.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 59.42499923706055, "explorer_subspace_norm": 117.07512664794922, "final_distance": 4.730400562286377, "max_param_abs": 1.606895923614502, "policy_loss": -0.0002411644566997495, "value_loss": 3.523293375968933}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.606931+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.116372585296631, "chaser_new_fields": 76.8699951171875, "chaser_partner_vision": 0.0, "chaser_return": 5.221996784210205, "chaser_stuck_run_steps": 0.9899999499320984, "collisions_per_episode": 2.9099998474121094, "degenerate_episodes": 87, "degenerate_fraction": 0.87, "episodes": 100, "explorer_new_fields": 32.80999755859375, "explorer_partner_vision": 0.0, "explorer_return": -1.5199999809265137, "explorer_stuck_run_steps": 2.7300000190734863, "final_distance": 4.800493240356445, "same_state_run_steps": 0.09999999403953552}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0005"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00022429173202174373, "chaser_grad_norm": 0.04249562695622444, "chaser_new_fields": 80.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.015000820159912, "chaser_subspace_norm": 93.73216247558594, "collisions_per_episode": 1.024999976158142, "entropy": 0.05529557354748249, "explorer_grad_norm": 0.39152009785175323, "explorer_new_fields": 73.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 59.42499923706055, "explorer_subspace_norm": 117.07512664794922, "final_distance": 4.730400562286377, "max_param_abs": 1.606895923614502, "policy_loss": -0.0002411644566997495, "value_loss": 3.523293375968933}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.703008+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.121584415435791, "chaser_new_fields": 33.97999954223633, "chaser_partner_vision": 0.0, "chaser_return": -3.2599987983703613, "chaser_stuck_run_steps": 2.629999876022339, "collisions_per_episode": 2.3499999046325684, "degenerate_episodes": 97, "degenerate_fraction": 0.97, "episodes": 100, "explorer_new_fields": 71.93999481201172, "explorer_partner_vision": 0.0, "explorer_return": 57.11499786376953, "explorer_stuck_run_steps": 1.559999942779541, "final_distance": 4.933297634124756, "same_state_run_steps": 0.11999999731779099}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0005"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 8.635803430578903e-06, "chaser_grad_norm": 0.10996681079268456, "chaser_new_fields": 82.375, "chaser_partner_vision": 0.0, "chaser_return": 6.390000820159912, "chaser_subspace_norm": 73.99776458740234, "collisions_per_episode": 1.149999976158142, "entropy": 0.05159799661487341, "explorer_grad_norm": 0.7565127648413181, "explorer_new_fields": 61.025001525878906, "explorer_partner_vision": 0.0, "explorer_return": 41.23749923706055, "explorer_subspace_norm": 46.546077728271484, "final_distance": 4.668344020843506, "max_param_abs": 1.2401388883590698, "policy_loss": -0.0002673804239634592, "value_loss": 3.8410890698432922}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.802941+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.091162204742432, "chaser_new_fields": 77.73999786376953, "chaser_partner_vision": 0.0, "chaser_return": 5.407996654510498, "chaser_stuck_run_steps": 0.09999999403953552, "collisions_per_episode": 2.509999990463257, "degenerate_episodes": 86, "degenerate_fraction": 0.86, "episodes": 100, "explorer_new_fields": 32.80999755859375, "explorer_partner_vision": 0.0, "explorer_return": -1.3399999141693115, "explorer_stuck_run_steps": 2.700000047683716, "final_distance": 5.153555870056152, "same_state_run_steps": 0.019999999552965164}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0006"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 8.635803430578903e-06, "chaser_grad_norm": 0.10996681079268456, "chaser_new_fields": 82.375, "chaser_partner_vision": 0.0, "chaser_return": 6.390000820159912, "chaser_subspace_norm": 73.99776458740234, "collisions_per_episode": 1.149999976158142, "entropy": 0.05159799661487341, "explorer_grad_norm": 0.7565127648413181, "explorer_new_fields": 61.025001525878906, "explorer_partner_vision": 0.0, "explorer_return": 41.23749923706055, "explorer_subspace_norm": 46.546077728271484, "final_distance": 4.668344020843506, "max_param_abs": 1.2401388883590698, "policy_loss": -0.0002673804239634592, "value_loss": 3.8410890698432922}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:43.900898+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.003658771514893, "chaser_new_fields": 33.75, "chaser_partner_vision": 0.0, "chaser_return": -3.3079988956451416, "chaser_stuck_run_steps": 2.629999876022339, "collisions_per_episode": 2.4800000190734863, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 60.68000030517578, "explorer_partner_vision": 0.0, "explorer_return": 40.404998779296875, "explorer_stuck_run_steps": 0.3199999928474426, "final_distance": 4.824435710906982, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0006"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005693047922270544, "chaser_grad_norm": 0.08362068422138691, "chaser_new_fields": 82.0250015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.3000006675720215, "chaser_subspace_norm": 109.71160888671875, "collisions_per_episode": 2.174999952316284, "entropy": 0.05865261983126402, "explorer_grad_norm": 0.6362885981798172, "explorer_new_fields": 60.57500076293945, "explorer_partner_vision": 0.0, "explorer_return": 40.07500076293945, "explorer_subspace_norm": 35.61513900756836, "final_distance": 4.870059967041016, "max_param_abs": 1.4202895164489746, "policy_loss": -0.0008471746835780891, "value_loss": 4.014994144439697}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.001633+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.128373146057129, "chaser_new_fields": 82.11000061035156, "chaser_partner_vision": 0.0, "chaser_return": 6.269996166229248, "chaser_stuck_run_steps": 0.08999999612569809, "collisions_per_episode": 2.93999981880188, "degenerate_episodes": 86, "degenerate_fraction": 0.86, "episodes": 100, "explorer_new_fields": 33.040000915527344, "explorer_partner_vision": 0.0, "explorer_return": -0.9799999594688416, "explorer_stuck_run_steps": 2.7200000286102295, "final_distance": 5.028230667114258, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0007"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005693047922270544, "chaser_grad_norm": 0.08362068422138691, "chaser_new_fields": 82.0250015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.3000006675720215, "chaser_subspace_norm": 109.71160888671875, "collisions_per_episode": 2.174999952316284, "entropy": 0.05865261983126402, "explorer_grad_norm": 0.6362885981798172, "explorer_new_fields": 60.57500076293945, "explorer_partner_vision": 0.0, "explorer_return": 40.07500076293945, "explorer_subspace_norm": 35.61513900756836, "final_distance": 4.870059967041016, "max_param_abs": 1.4202895164489746, "policy_loss": -0.0008471746835780891, "value_loss": 4.014994144439697}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.097744+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.059652328491211, "chaser_new_fields": 33.790000915527344, "chaser_partner_vision": 0.0, "chaser_return": -3.2999987602233887, "chaser_stuck_run_steps": 2.6499998569488525, "collisions_per_episode": 2.2899999618530273, "degenerate_episodes": 90, "degenerate_fraction": 0.9, "episodes": 100, "explorer_new_fields": 61.14999771118164, "explorer_partner_vision": 0.0, "explorer_return": 41.125, "explorer_stuck_run_steps": 0.19999998807907104, "final_distance": 4.463276386260986, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0007"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 3.927184380664306e-06, "chaser_grad_norm": 0.07202051393687725, "chaser_new_fields": 81.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.145000457763672, "chaser_subspace_norm": 132.7053680419922, "collisions_per_episode": 1.975000023841858, "entropy": 0.13418130949139595, "explorer_grad_norm": 0.6797178164124489, "explorer_new_fields": 68.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 52.48749923706055, "explorer_subspace_norm": 61.86991500854492, "final_distance": 4.521096229553223, "max_param_abs": 1.4186153411865234, "policy_loss": -0.0009259324722812323, "value_loss": 4.002953112125397}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.197891+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.262848854064941, "chaser_new_fields": 78.5, "chaser_partner_vision": 0.0, "chaser_return": 5.547996520996094, "chaser_stuck_run_steps": 0.03999999910593033, "collisions_per_episode": 2.440000057220459, "degenerate_episodes": 83, "degenerate_fraction": 0.83, "episodes": 100, "explorer_new_fields": 32.88999938964844, "explorer_partner_vision": 0.0, "explorer_return": -1.3700000047683716, "explorer_stuck_run_steps": 2.740000009536743, "final_distance": 5.147022724151611, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0008"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 3.927184380664306e-06, "chaser_grad_norm": 0.07202051393687725, "chaser_new_fields": 81.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.145000457763672, "chaser_subspace_norm": 132.7053680419922, "collisions_per_episode": 1.975000023841858, "entropy": 0.13418130949139595, "explorer_grad_norm": 0.6797178164124489, "explorer_new_fields": 68.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 52.48749923706055, "explorer_subspace_norm": 61.86991500854492, "final_distance": 4.521096229553223, "max_param_abs": 1.4186153411865234, "policy_loss": -0.0009259324722812323, "value_loss": 4.002953112125397}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.294102+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.165109157562256, "chaser_new_fields": 33.48999786376953, "chaser_partner_vision": 0.0, "chaser_return": -3.343998908996582, "chaser_stuck_run_steps": 2.5799999237060547, "collisions_per_episode": 2.2699999809265137, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 68.97000122070312, "explorer_partner_vision": 0.0, "explorer_return": 52.70499801635742, "explorer_stuck_run_steps": 0.8499999642372131, "final_distance": 5.33642578125, "same_state_run_steps": 0.12999999523162842}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0008"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00014854547405664803, "chaser_grad_norm": 0.09859942272305489, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.340001106262207, "chaser_subspace_norm": 98.81183624267578, "collisions_per_episode": 2.7250001430511475, "entropy": 0.07547229155898094, "explorer_grad_norm": 0.32836734503507614, "explorer_new_fields": 65.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 47.462501525878906, "explorer_subspace_norm": 26.826025009155273, "final_distance": 5.683727741241455, "max_param_abs": 1.2476426362991333, "policy_loss": -0.0011177757633866747, "value_loss": 3.991724371910095}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.394659+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 5.134922504425049, "chaser_new_fields": 82.80999755859375, "chaser_partner_vision": 0.0, "chaser_return": 6.391995906829834, "chaser_stuck_run_steps": 0.009999999776482582, "collisions_per_episode": 2.819999933242798, "degenerate_episodes": 83, "degenerate_fraction": 0.83, "episodes": 100, "explorer_new_fields": 32.959999084472656, "explorer_partner_vision": 0.0, "explorer_return": -1.2649999856948853, "explorer_stuck_run_steps": 2.679999828338623, "final_distance": 5.392402648925781, "same_state_run_steps": 0.0}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0009"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00014854547405664803, "chaser_grad_norm": 0.09859942272305489, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.340001106262207, "chaser_subspace_norm": 98.81183624267578, "collisions_per_episode": 2.7250001430511475, "entropy": 0.07547229155898094, "explorer_grad_norm": 0.32836734503507614, "explorer_new_fields": 65.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 47.462501525878906, "explorer_subspace_norm": 26.826025009155273, "final_distance": 5.683727741241455, "max_param_abs": 1.2476426362991333, "policy_loss": -0.0011177757633866747, "value_loss": 3.991724371910095}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.491290+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 4.98706579208374, "chaser_new_fields": 33.81999969482422, "chaser_partner_vision": 0.0, "chaser_return": -3.3119988441467285, "chaser_stuck_run_steps": 2.619999885559082, "collisions_per_episode": 2.490000009536743, "degenerate_episodes": 86, "degenerate_fraction": 0.86, "episodes": 100, "explorer_new_fields": 64.93999481201172, "explorer_partner_vision": 0.0, "explorer_return": 46.67499923706055, "explorer_stuck_run_steps": 0.029999999329447746, "final_distance": 4.802732467651367, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0009"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.002805363851033249, "chaser_grad_norm": 0.188689436763525, "chaser_new_fields": 52.92499923706055, "chaser_partner_vision": 0.9490000605583191, "chaser_return": 23.260000228881836, "chaser_subspace_norm": 46.499481201171875, "collisions_per_episode": 21.450000762939453, "entropy": 0.1341038942337036, "explorer_grad_norm": 1.1695344895124435, "explorer_new_fields": 65.55000305175781, "explorer_partner_vision": 0.9490000605583191, "explorer_return": 31.037500381469727, "explorer_subspace_norm": 105.87859344482422, "final_distance": 1.655848741531372, "max_param_abs": 1.2022045850753784, "policy_loss": -0.003802475623324142, "value_loss": 4.788339614868164}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.600352+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.9226698875427246, "chaser_new_fields": 31.739999771118164, "chaser_partner_vision": 0.9387999773025513, "chaser_return": 22.954994201660156, "chaser_stuck_run_steps": 0.6100000143051147, "collisions_per_episode": 24.729999542236328, "degenerate_episodes": 94, "degenerate_fraction": 0.94, "episodes": 100, "explorer_new_fields": 27.779998779296875, "explorer_partner_vision": 0.9387999773025513, "explorer_return": -26.604999542236328, "explorer_stuck_run_steps": 3.259999990463257, "final_distance": 1.7273223400115967, "same_state_run_steps": 0.26999998092651367}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0000"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.002805363851033249, "chaser_grad_norm": 0.188689436763525, "chaser_new_fields": 52.92499923706055, "chaser_partner_vision": 0.9490000605583191, "chaser_return": 23.260000228881836, "chaser_subspace_norm": 46.499481201171875, "collisions_per_episode": 21.450000762939453, "entropy": 0.1341038942337036, "explorer_grad_norm": 1.1695344895124435, "explorer_new_fields": 65.55000305175781, "explorer_partner_vision": 0.9490000605583191, "explorer_return": 31.037500381469727, "explorer_subspace_norm": 105.87859344482422, "final_distance": 1.655848741531372, "max_param_abs": 1.2022045850753784, "policy_loss": -0.003802475623324142, "value_loss": 4.788339614868164}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.704418+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.418818473815918, "chaser_new_fields": 33.89999771118164, "chaser_partner_vision": 0.2913999855518341, "chaser_return": -1.88499915599823, "chaser_stuck_run_steps": 2.7100000381469727, "collisions_per_episode": 1.25, "degenerate_episodes": 91, "degenerate_fraction": 0.91, "episodes": 100, "explorer_new_fields": 55.38999938964844, "explorer_partner_vision": 0.2913999855518341, "explorer_return": 32.06999969482422, "explorer_stuck_run_steps": 0.12999999523162842, "final_distance": 5.25391149520874, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0000"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005562261293015069, "chaser_grad_norm": 0.6423081681132317, "chaser_new_fields": 46.20000076293945, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 23.4325008392334, "chaser_subspace_norm": 57.10129928588867, "collisions_per_episode": 22.625, "entropy": 0.14692536741495132, "explorer_grad_norm": 0.5817638859152794, "explorer_new_fields": 52.775001525878906, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 12.71250057220459, "explorer_subspace_norm": 122.46495819091797, "final_distance": 1.791115403175354, "max_param_abs": 1.3355721235275269, "policy_loss": -0.004476253691127141, "value_loss": 5.1513988971710205}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.812676+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 2.0339229106903076, "chaser_new_fields": 33.209999084472656, "chaser_partner_vision": 0.9183999300003052, "chaser_return": 21.927995681762695, "chaser_stuck_run_steps": 0.6699999570846558, "collisions_per_episode": 23.51999855041504, "degenerate_episodes": 92, "degenerate_fraction": 0.92, "episodes": 100, "explorer_new_fields": 26.84000015258789, "explorer_partner_vision": 0.9183999300003052, "explorer_return": -26.510000228881836, "explorer_stuck_run_steps": 3.169999837875366, "final_distance": 1.7653462886810303, "same_state_run_steps": 0.25}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0001"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005562261293015069, "chaser_grad_norm": 0.6423081681132317, "chaser_new_fields": 46.20000076293945, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 23.4325008392334, "chaser_subspace_norm": 57.10129928588867, "collisions_per_episode": 22.625, "entropy": 0.14692536741495132, "explorer_grad_norm": 0.5817638859152794, "explorer_new_fields": 52.775001525878906, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 12.71250057220459, "explorer_subspace_norm": 122.46495819091797, "final_distance": 1.791115403175354, "max_param_abs": 1.3355721235275269, "policy_loss": -0.004476253691127141, "value_loss": 5.1513988971710205}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:44.916857+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.981630325317383, "chaser_new_fields": 33.93000030517578, "chaser_partner_vision": 0.2247999906539917, "chaser_return": -2.4529991149902344, "chaser_stuck_run_steps": 2.690000057220459, "collisions_per_episode": 0.7099999785423279, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 53.73999786376953, "explorer_partner_vision": 0.2247999906539917, "explorer_return": 29.98499870300293, "explorer_stuck_run_steps": 0.28999999165534973, "final_distance": 5.55134916305542, "same_state_run_steps": 0.03999999910593033}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0001"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0014266868641836794, "chaser_grad_norm": 0.20290611311793327, "chaser_new_fields": 49.29999923706055, "chaser_partner_vision": 0.9287500381469727, "chaser_return": 18.48750114440918, "chaser_subspace_norm": 48.7114143371582, "collisions_per_episode": 17.625, "entropy": 0.18204283714294434, "explorer_grad_norm": 0.6118505597114563, "explorer_new_fields": 60.95000076293945, "explorer_partner_vision": 0.9287500381469727, "explorer_return": 26.837499618530273, "explorer_subspace_norm": 104.41957092285156, "final_distance": 1.6572967767715454, "max_param_abs": 1.1167881488800049, "policy_loss": -0.005259661337355226, "value_loss": 5.285007953643799}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.025324+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.846373200416565, "chaser_new_fields": 30.559999465942383, "chaser_partner_vision": 0.9455000162124634, "chaser_return": 25.47699737548828, "chaser_stuck_run_steps": 2.4099998474121094, "collisions_per_episode": 27.189998626708984, "degenerate_episodes": 95, "degenerate_fraction": 0.95, "episodes": 100, "explorer_new_fields": 25.549999237060547, "explorer_partner_vision": 0.9455000162124634, "explorer_return": -29.889999389648438, "explorer_stuck_run_steps": 3.359999895095825, "final_distance": 1.641021728515625, "same_state_run_steps": 0.5399999618530273}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0002"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0014266868641836794, "chaser_grad_norm": 0.20290611311793327, "chaser_new_fields": 49.29999923706055, "chaser_partner_vision": 0.9287500381469727, "chaser_return": 18.48750114440918, "chaser_subspace_norm": 48.7114143371582, "collisions_per_episode": 17.625, "entropy": 0.18204283714294434, "explorer_grad_norm": 0.6118505597114563, "explorer_new_fields": 60.95000076293945, "explorer_partner_vision": 0.9287500381469727, "explorer_return": 26.837499618530273, "explorer_subspace_norm": 104.41957092285156, "final_distance": 1.6572967767715454, "max_param_abs": 1.1167881488800049, "policy_loss": -0.005259661337355226, "value_loss": 5.285007953643799}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.131142+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.628232479095459, "chaser_new_fields": 33.95000076293945, "chaser_partner_vision": 0.26339998841285706, "chaser_return": -2.1799991130828857, "chaser_stuck_run_steps": 2.629999876022339, "collisions_per_episode": 0.9599999785423279, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 54.0, "explorer_partner_vision": 0.26339998841285706, "explorer_return": 30.189998626708984, "explorer_stuck_run_steps": 0.019999999552965164, "final_distance": 5.8266167640686035, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0002"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.000801136315581541, "chaser_grad_norm": 0.38031933456659317, "chaser_new_fields": 52.5, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 21.880001068115234, "chaser_subspace_norm": 52.41925811767578, "collisions_per_episode": 20.200000762939453, "entropy": 0.12773804366588593, "explorer_grad_norm": 0.23243309557437897, "explorer_new_fields": 59.29999923706055, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 22.887500762939453, "explorer_subspace_norm": 119.99281311035156, "final_distance": 1.4895528554916382, "max_param_abs": 1.4659936428070068, "policy_loss": -0.0027591757117768623, "value_loss": 5.230631113052368}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.255109+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.8304731845855713, "chaser_new_fields": 30.989999771118164, "chaser_partner_vision": 0.9462000131607056, "chaser_return": 27.23999786376953, "chaser_stuck_run_steps": 0.85999995470047, "collisions_per_episode": 28.760000228881836, "degenerate_episodes": 98, "degenerate_fraction": 0.98, "episodes": 100, "explorer_new_fields": 25.8799991607666, "explorer_partner_vision": 0.9462000131607056, "explorer_return": -31.03499984741211, "explorer_stuck_run_steps": 3.3499999046325684, "final_distance": 1.5262326002120972, "same_state_run_steps": 0.3799999952316284}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0003"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.000801136315581541, "chaser_grad_norm": 0.38031933456659317, "chaser_new_fields": 52.5, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 21.880001068115234, "chaser_subspace_norm": 52.41925811767578, "collisions_per_episode": 20.200000762939453, "entropy": 0.12773804366588593, "explorer_grad_norm": 0.23243309557437897, "explorer_new_fields": 59.29999923706055, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 22.887500762939453, "explorer_subspace_norm": 119.99281311035156, "final_distance": 1.4895528554916382, "max_param_abs": 1.4659936428070068, "policy_loss": -0.0027591757117768623, "value_loss": 5.230631113052368}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.376361+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.968653202056885, "chaser_new_fields": 33.86000061035156, "chaser_partner_vision": 0.22360000014305115, "chaser_return": -2.4379990100860596, "chaser_stuck_run_steps": 2.6599998474121094, "collisions_per_episode": 0.7400000095367432, "degenerate_episodes": 91, "degenerate_fraction": 0.91, "episodes": 100, "explorer_new_fields": 55.52000045776367, "explorer_partner_vision": 0.22360000014305115, "explorer_return": 32.59499740600586, "explorer_stuck_run_steps": 0.17999999225139618, "final_distance": 5.932736873626709, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0003"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.000236154562574864, "chaser_grad_norm": 0.38722849637269974, "chaser_new_fields": 45.57500076293945, "chaser_partner_vision": 0.9225000143051147, "chaser_return": 19.41750144958496, "chaser_subspace_norm": 58.249393463134766, "collisions_per_episode": 19.125, "entropy": 0.19220001995563507, "explorer_grad_norm": 0.3825383335351944, "explorer_new_fields": 56.85000228881836, "explorer_partner_vision": 0.9225000143051147, "explorer_return": 19.30000114440918, "explorer_subspace_norm": 90.54417419433594, "final_distance": 1.830224633216858, "max_param_abs": 1.0662928819656372, "policy_loss": -0.004796283046713867, "value_loss": 5.130034923553467}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.500759+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.7290464639663696, "chaser_new_fields": 28.920000076293945, "chaser_partner_vision": 0.9477999806404114, "chaser_return": 36.003997802734375, "chaser_stuck_run_steps": 1.3799999952316284, "collisions_per_episode": 37.13999938964844, "degenerate_episodes": 97, "degenerate_fraction": 0.97, "episodes": 100, "explorer_new_fields": 23.78999900817871, "explorer_partner_vision": 0.9477999806404114, "explorer_return": -39.125, "explorer_stuck_run_steps": 3.5999999046325684, "final_distance": 1.566463589668274, "same_state_run_steps": 0.5299999713897705}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0004"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.000236154562574864, "chaser_grad_norm": 0.38722849637269974, "chaser_new_fields": 45.57500076293945, "chaser_partner_vision": 0.9225000143051147, "chaser_return": 19.41750144958496, "chaser_subspace_norm": 58.249393463134766, "collisions_per_episode": 19.125, "entropy": 0.19220001995563507, "explorer_grad_norm": 0.3825383335351944, "explorer_new_fields": 56.85000228881836, "explorer_partner_vision": 0.9225000143051147, "explorer_return": 19.30000114440918, "explorer_subspace_norm": 90.54417419433594, "final_distance": 1.830224633216858, "max_param_abs": 1.0662928819656372, "policy_loss": -0.004796283046713867, "value_loss": 5.130034923553467}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.610487+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 6.055323123931885, "chaser_new_fields": 33.90999984741211, "chaser_partner_vision": 0.204599991440773, "chaser_return": -2.4979991912841797, "chaser_stuck_run_steps": 2.619999885559082, "collisions_per_episode": 0.6800000071525574, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 46.849998474121094, "explorer_partner_vision": 0.204599991440773, "explorer_return": 19.69499969482422, "explorer_stuck_run_steps": 0.07000000029802322, "final_distance": 6.047131061553955, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0004"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005827252787078407, "chaser_grad_norm": 0.3728112056851387, "chaser_new_fields": 51.47500228881836, "chaser_partner_vision": 0.95250004529953, "chaser_return": 20.117502212524414, "chaser_subspace_norm": 50.84516143798828, "collisions_per_episode": 18.575000762939453, "entropy": 0.13566195219755173, "explorer_grad_norm": 0.6062372475862503, "explorer_new_fields": 58.57500076293945, "explorer_partner_vision": 0.95250004529953, "explorer_return": 20.774999618530273, "explorer_subspace_norm": 104.22787475585938, "final_distance": 1.5808099508285522, "max_param_abs": 1.3590010404586792, "policy_loss": -0.003127762281341262, "value_loss": 4.4600828886032104}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.725728+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 2.022212505340576, "chaser_new_fields": 31.869998931884766, "chaser_partner_vision": 0.9148999452590942, "chaser_return": 23.19499397277832, "chaser_stuck_run_steps": 0.85999995470047, "collisions_per_episode": 24.889999389648438, "degenerate_episodes": 92, "degenerate_fraction": 0.92, "episodes": 100, "explorer_new_fields": 27.619998931884766, "explorer_partner_vision": 0.9148999452590942, "explorer_return": -25.889999389648438, "explorer_stuck_run_steps": 3.240000009536743, "final_distance": 1.8503373861312866, "same_state_run_steps": 0.25999999046325684}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0005"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0005827252787078407, "chaser_grad_norm": 0.3728112056851387, "chaser_new_fields": 51.47500228881836, "chaser_partner_vision": 0.95250004529953, "chaser_return": 20.117502212524414, "chaser_subspace_norm": 50.84516143798828, "collisions_per_episode": 18.575000762939453, "entropy": 0.13566195219755173, "explorer_grad_norm": 0.6062372475862503, "explorer_new_fields": 58.57500076293945, "explorer_partner_vision": 0.95250004529953, "explorer_return": 20.774999618530273, "explorer_subspace_norm": 104.22787475585938, "final_distance": 1.5808099508285522, "max_param_abs": 1.3590010404586792, "policy_loss": -0.003127762281341262, "value_loss": 4.4600828886032104}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.829596+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 6.011486530303955, "chaser_new_fields": 33.73999786376953, "chaser_partner_vision": 0.2134999781847, "chaser_return": -2.4539990425109863, "chaser_stuck_run_steps": 2.6599998474121094, "collisions_per_episode": 0.7400000095367432, "degenerate_episodes": 91, "degenerate_fraction": 0.91, "episodes": 100, "explorer_new_fields": 43.37999725341797, "explorer_partner_vision": 0.2134999781847, "explorer_return": 14.429999351501465, "explorer_stuck_run_steps": 0.17000000178813934, "final_distance": 5.857645034790039, "same_state_run_steps": 0.019999999552965164}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0005"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00025015939701367174, "chaser_grad_norm": 0.31481391191482544, "chaser_new_fields": 49.900001525878906, "chaser_partner_vision": 0.9252500534057617, "chaser_return": 21.020002365112305, "chaser_subspace_norm": 49.21955108642578, "collisions_per_episode": 19.80000114440918, "entropy": 0.1560382954776287, "explorer_grad_norm": 0.6951030343770981, "explorer_new_fields": 59.42499923706055, "explorer_partner_vision": 0.9252500534057617, "explorer_return": 22.48750114440918, "explorer_subspace_norm": 113.72798919677734, "final_distance": 1.6350990533828735, "max_param_abs": 1.6354095935821533, "policy_loss": -0.0035889741121788443, "value_loss": 4.686129331588745}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:45.937143+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.8730149269104004, "chaser_new_fields": 31.01999855041504, "chaser_partner_vision": 0.939300000667572, "chaser_return": 26.81099510192871, "chaser_stuck_run_steps": 0.7799999713897705, "collisions_per_episode": 28.369998931884766, "degenerate_episodes": 97, "degenerate_fraction": 0.97, "episodes": 100, "explorer_new_fields": 25.760000228881836, "explorer_partner_vision": 0.939300000667572, "explorer_return": -30.899999618530273, "explorer_stuck_run_steps": 3.3299999237060547, "final_distance": 1.75227952003479, "same_state_run_steps": 0.2800000011920929}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0006"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00025015939701367174, "chaser_grad_norm": 0.31481391191482544, "chaser_new_fields": 49.900001525878906, "chaser_partner_vision": 0.9252500534057617, "chaser_return": 21.020002365112305, "chaser_subspace_norm": 49.21955108642578, "collisions_per_episode": 19.80000114440918, "entropy": 0.1560382954776287, "explorer_grad_norm": 0.6951030343770981, "explorer_new_fields": 59.42499923706055, "explorer_partner_vision": 0.9252500534057617, "explorer_return": 22.48750114440918, "explorer_subspace_norm": 113.72798919677734, "final_distance": 1.6350990533828735, "max_param_abs": 1.6354095935821533, "policy_loss": -0.0035889741121788443, "value_loss": 4.686129331588745}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.040804+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.9493937492370605, "chaser_new_fields": 33.93000030517578, "chaser_partner_vision": 0.21109998226165771, "chaser_return": -2.5089991092681885, "chaser_stuck_run_steps": 2.609999895095825, "collisions_per_episode": 0.6699999570846558, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 53.27000045776367, "explorer_partner_vision": 0.21109998226165771, "explorer_return": 29.344999313354492, "explorer_stuck_run_steps": 0.07000000029802322, "final_distance": 5.721303462982178, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0006"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00013653355120357702, "chaser_grad_norm": 0.9264960587024689, "chaser_new_fields": 46.92499923706055, "chaser_partner_vision": 0.9470000267028809, "chaser_return": 24.285001754760742, "chaser_subspace_norm": 67.47857666015625, "collisions_per_episode": 23.200000762939453, "entropy": 0.1407805271446705, "explorer_grad_norm": 0.5267750471830368, "explorer_new_fields": 55.900001525878906, "explorer_partner_vision": 0.9470000267028809, "explorer_return": 17.149999618530273, "explorer_subspace_norm": 89.71617889404297, "final_distance": 1.689988374710083, "max_param_abs": 1.2198518514633179, "policy_loss": -0.003363337086513596, "value_loss": 4.716352343559265}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.157423+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.8591976165771484, "chaser_new_fields": 30.69999885559082, "chaser_partner_vision": 0.9456999897956848, "chaser_return": 26.13799285888672, "chaser_stuck_run_steps": 1.0299999713897705, "collisions_per_episode": 27.799999237060547, "degenerate_episodes": 94, "degenerate_fraction": 0.94, "episodes": 100, "explorer_new_fields": 26.239999771118164, "explorer_partner_vision": 0.9456999897956848, "explorer_return": -29.90999984741211, "explorer_stuck_run_steps": 3.369999885559082, "final_distance": 1.777151107788086, "same_state_run_steps": 0.4899999797344208}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0007"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.00013653355120357702, "chaser_grad_norm": 0.9264960587024689, "chaser_new_fields": 46.92499923706055, "chaser_partner_vision": 0.9470000267028809, "chaser_return": 24.285001754760742, "chaser_subspace_norm": 67.47857666015625, "collisions_per_episode": 23.200000762939453, "entropy": 0.1407805271446705, "explorer_grad_norm": 0.5267750471830368, "explorer_new_fields": 55.900001525878906, "explorer_partner_vision": 0.9470000267028809, "explorer_return": 17.149999618530273, "explorer_subspace_norm": 89.71617889404297, "final_distance": 1.689988374710083, "max_param_abs": 1.2198518514633179, "policy_loss": -0.003363337086513596, "value_loss": 4.716352343559265}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.263149+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.806032657623291, "chaser_new_fields": 33.84000015258789, "chaser_partner_vision": 0.23879998922348022, "chaser_return": -2.4399991035461426, "chaser_stuck_run_steps": 2.6499998569488525, "collisions_per_episode": 0.7400000095367432, "degenerate_episodes": 89, "degenerate_fraction": 0.89, "episodes": 100, "explorer_new_fields": 57.55999755859375, "explorer_partner_vision": 0.23879998922348022, "explorer_return": 35.744998931884766, "explorer_stuck_run_steps": 0.08999999612569809, "final_distance": 6.035492897033691, "same_state_run_steps": 0.0}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0007"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0022941239069333985, "chaser_grad_norm": 0.43133843690156937, "chaser_new_fields": 50.125, "chaser_partner_vision": 0.9205000400543213, "chaser_return": 20.852500915527344, "chaser_subspace_norm": 52.702186584472656, "collisions_per_episode": 19.67500114440918, "entropy": 0.15483620017766953, "explorer_grad_norm": 0.3490636646747589, "explorer_new_fields": 60.525001525878906, "explorer_partner_vision": 0.9205000400543213, "explorer_return": 23.0, "explorer_subspace_norm": 101.96681213378906, "final_distance": 1.8305482864379883, "max_param_abs": 1.1802449226379395, "policy_loss": -0.00493374926503698, "value_loss": 4.697651386260986}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.371355+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 1.9622390270233154, "chaser_new_fields": 31.35999870300293, "chaser_partner_vision": 0.9269000291824341, "chaser_return": 23.311994552612305, "chaser_stuck_run_steps": 0.6299999952316284, "collisions_per_episode": 25.100000381469727, "degenerate_episodes": 97, "degenerate_fraction": 0.97, "episodes": 100, "explorer_new_fields": 27.68000030517578, "explorer_partner_vision": 0.9269000291824341, "explorer_return": -26.309999465942383, "explorer_stuck_run_steps": 3.1399998664855957, "final_distance": 1.7391915321350098, "same_state_run_steps": 0.20999999344348907}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0008"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": 0.0022941239069333985, "chaser_grad_norm": 0.43133843690156937, "chaser_new_fields": 50.125, "chaser_partner_vision": 0.9205000400543213, "chaser_return": 20.852500915527344, "chaser_subspace_norm": 52.702186584472656, "collisions_per_episode": 19.67500114440918, "entropy": 0.15483620017766953, "explorer_grad_norm": 0.3490636646747589, "explorer_new_fields": 60.525001525878906, "explorer_partner_vision": 0.9205000400543213, "explorer_return": 23.0, "explorer_subspace_norm": 101.96681213378906, "final_distance": 1.8305482864379883, "max_param_abs": 1.1802449226379395, "policy_loss": -0.00493374926503698, "value_loss": 4.697651386260986}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.475135+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.686277866363525, "chaser_new_fields": 33.880001068115234, "chaser_partner_vision": 0.2646999955177307, "chaser_return": -2.040999412536621, "chaser_stuck_run_steps": 2.609999895095825, "collisions_per_episode": 1.090000033378601, "degenerate_episodes": 88, "degenerate_fraction": 0.88, "episodes": 100, "explorer_new_fields": 55.25, "explorer_partner_vision": 0.2646999955177307, "explorer_return": 31.864999771118164, "explorer_stuck_run_steps": 0.26999998092651367, "final_distance": 5.499191761016846, "same_state_run_steps": 0.009999999776482582}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0008"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": -0.0002899035942244277, "chaser_grad_norm": 0.2534610331058502, "chaser_new_fields": 51.625, "chaser_partner_vision": 0.9552500247955322, "chaser_return": 22.76500129699707, "chaser_subspace_norm": 57.69636535644531, "collisions_per_episode": 21.25, "entropy": 0.1398237645626068, "explorer_grad_norm": 0.3390047997236252, "explorer_new_fields": 58.42499923706055, "explorer_partner_vision": 0.9552500247955322, "explorer_return": 21.6875, "explorer_subspace_norm": 105.33492279052734, "final_distance": 1.5442389249801636, "max_param_abs": 1.1976157426834106, "policy_loss": -0.002912267684005876, "value_loss": 4.631349444389343}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.583483+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "chaser", "max_steps": 100, "metrics": {"average_distance": 2.13496994972229, "chaser_new_fields": 33.89999771118164, "chaser_partner_vision": 0.8927000164985657, "chaser_return": 20.284996032714844, "chaser_stuck_run_steps": 0.550000011920929, "collisions_per_episode": 21.850000381469727, "degenerate_episodes": 94, "degenerate_fraction": 0.94, "episodes": 100, "explorer_new_fields": 28.149999618530273, "explorer_partner_vision": 0.8927000164985657, "explorer_return": -23.260000228881836, "explorer_stuck_run_steps": 3.2200000286102295, "final_distance": 1.8080952167510986, "same_state_run_steps": 0.25}, "opponent_mode": "random_explorer", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0009"}
{"attempt_id": "attempt_01", "batch_size": 128, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_metrics": {"approx_kl": -0.0002899035942244277, "chaser_grad_norm": 0.2534610331058502, "chaser_new_fields": 51.625, "chaser_partner_vision": 0.9552500247955322, "chaser_return": 22.76500129699707, "chaser_subspace_norm": 57.69636535644531, "collisions_per_episode": 21.25, "entropy": 0.1398237645626068, "explorer_grad_norm": 0.3390047997236252, "explorer_new_fields": 58.42499923706055, "explorer_partner_vision": 0.9552500247955322, "explorer_return": 21.6875, "explorer_subspace_norm": 105.33492279052734, "final_distance": 1.5442389249801636, "max_param_abs": 1.1976157426834106, "policy_loss": -0.002912267684005876, "value_loss": 4.631349444389343}, "command": "/root/mouse-run-run-perf/.venv/bin/evaluate-paper-marl runs/mouse-run-run-2-mlp --output runs/mouse-run-run-2-mlp/paper_random_eval.jsonl --device cuda", "created_at": "2026-07-02T11:50:46.687176+00:00", "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episodes": 100, "error": null, "evaluation_protocol": "paper_random_opponent_v1", "experiment_id": "mouse-run-run-2-mlp", "focus_agent": "explorer", "max_steps": 100, "metrics": {"average_distance": 5.822822093963623, "chaser_new_fields": 33.88999938964844, "chaser_partner_vision": 0.2704000174999237, "chaser_return": -1.4439992904663086, "chaser_stuck_run_steps": 2.609999895095825, "collisions_per_episode": 1.659999966621399, "degenerate_episodes": 87, "degenerate_fraction": 0.87, "episodes": 100, "explorer_new_fields": 49.029998779296875, "explorer_partner_vision": 0.2704000174999237, "explorer_return": 22.23499870300293, "explorer_stuck_run_steps": 0.14999999105930328, "final_distance": 6.008201599121094, "same_state_run_steps": 0.029999999329447746}, "opponent_mode": "random_chaser", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "schema_version": 3, "seed": 0, "status": "ok", "task": "social", "unit_id": "social/seed_0009"}